{"id":"aa12a612-e501-4fc8-9ce7-c9a514254268","arxiv_id":"2507.10187","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"On the surface of the Weyl semimetal t-PtBi2, a ~9 meV spectroscopic gap closes near 45 K, consistent with surface superconductivity at Tc ~ 50 K.","lead":"Scanning tunneling spectroscopy on the Weyl semimetal t-PtBi2 shows a roughly 9 meV energy gap on the crystal surface that shrinks as the sample is warmed and closes near 45 K, which the authors interpret as high-temperature surface superconductivity with Tc near 50 K. A smart generalist should care because a confirmed topological superconductor at this temperature would be a new phenomenon in condensed matter and a candidate platform for Majorana-based quantum computing.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The existence of surface superconductivity with Tc≈(50±5) K is not independently established: the gap-like dip is never shown to be a superconducting gap, and the supporting simulation in Fig. 3 assumes the claimed BCS ratio.","rationale":"The reader's conditional verdict is well calibrated. I read the raw observation as a genuine, reproducible fill-in of a spectroscopic dip with temperature; the authors are transparent about the FWHM-based gap estimate, the missing coherence peaks, and the 50 K spectrum's instability. Those features distinguish this from a misleading report. However, the chain from dI/dU dip to Tc≈50 K has one unguarded link: superconductivity is never independently confirmed. The paper's own discussion raises the pseudogap alternative, and the simulation in Fig. 3 cannot validate the interpretation because it is constructed from the BCS gap equation with the claimed BCS ratio. A field-dependent vortex test is the single most decisive check; it is technically feasible with the same STM and would separate a superconducting gap from correlation/pseudogap scenarios. If vortices are observed, the claim is substantially strengthened. If no vortices are observed and the dip persists to fields that should destroy a 9 meV surface gap, the Tc estimate must be reconsidered. Because the current evidence is suggestive but not sufficient, I keep the reader's CONDITIONAL verdict and see no basis to move to accept or reject.","tokens_in":8141,"tokens_out":4978,"duration_ms":62483,"concrete_test":"At T=8 K, perform field-dependent STS on the same Type-A surface: acquire ZBC maps over a 1–2 μm area for out-of-plane fields from 0 to several tesla. A superconducting surface should develop a vortex lattice (density B/Φ0) with enhanced zero-bias conductance at vortex cores and a dip that weakens toward Hc2; a pseudogap or density-wave gap would show no vortex pattern and would remain essentially field-insensitive on this scale. This test settles whether the spectroscopic dip is a superconducting gap without using the Fig. 3 simulation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the identification of the ~9 meV dI/dU dip as a superconducting gap. The raw temperature series in Fig. 2 shows a broad feature that progressively fills in, but nothing in the data forces the superconducting interpretation. The paper's own Discussion names incoherent pair formation and proximity to a quantum critical point as mechanisms that can produce a gap with suppressed coherence peaks, and the same paragraph notes the absence of the sharp coherence peaks seen in ARPES. The nonzero zero-bias conductance is attributed to the surface nature of the superconductivity without independent support. The Fig. 3 validation is not a check of the hypothesis: it assumes an s-wave Dynes DOS (Supplement Eq. S1) with Δ=9 meV, Γ=8±2 meV, and RBCS=4.2, thereby building in the very BCS ratio and Tc the paper claims to confirm. Finally, the 50 K spectrum, which would directly document the zero-gap normal state, is excluded from the main analysis because of a reported tunneling-junction change (Supplement Fig. S3); the estimate Tc≈(50±5) K therefore rests on an extrapolation from spectra that still retain a finite dip. If the feature is a pseudogap, density-wave gap, or a junction-related artifact, the estimated Tc, BCS ratio, and topological-superconductivity interpretation lose their evidentiary base.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports scanning tunneling spectroscopy (STS) measurements of the differential conductance dI/dU on the decorated honeycomb (Type A) surface of t-PtBi2 between 8 K and 45 K. The central observation is a broad zero-bias dip of ~38% depth at 8 K, with a half-width at half minimum of about 9 meV, which progressively fills in with increasing temperature faster than the purely thermal broadening of the 8 K spectrum. The authors interpret this as the closing of a superconducting gap and estimate Tc ≈ (50 ± 5) K and a BCS ratio 2Δ/kBTc ≈ 4.2 ± 0.2, concluding that intrinsic topological surface superconductivity with high Tc is confirmed. The paper includes a simulation of the zero-bias conductance based on a Dynes-broadened BCS density of states and a supplement documenting raw and processed spectra.","tokens_in":8311,"tokens_out":5750,"duration_ms":56619,"significance":"If the identification of the dip as a superconducting gap is correct, the claim that the surface of t-PtBi2 superconducts up to ~50 K, an order of magnitude above the bulk Tc ≈ 1 K, is extraordinary and would make t-PtBi2 the highest-Tc candidate for intrinsic topological surface superconductivity. The raw temperature series is a genuine asset: the comparison with the thermally broadened 8 K spectrum clearly shows that the dip fills in faster than trivial thermal broadening, providing model-independent evidence for a temperature-driven change in the electronic structure. The paper neither fits the spectra with a BCS form nor presents transport or magnetic evidence, so the quantitative claims (Tc, RBCS) rest on a single spectral feature and a simulation whose inputs are the claimed quantities. If the feature is a pseudogap, density-wave, or junction-related artifact, the central claims lose their evidentiary base.","major_comments":[{"comment":"The estimate Tc ≈ (50 ± 5) K is not directly supported by data: the 50 K spectrum, which would show the fully gapped-to-normal transition, is excluded from the main analysis because of a reported change of the tunneling junction, and at the highest consistently measured temperature of 45 K a finite (though strongly reduced) dip remains. The extrapolation from a shallow dip to a transition temperature therefore rests on an assumption, and the statement 'at T = 50 K the gap is fully closed' is based on an inconsistent spectrum. The Tc value and the BCS ratio RBCS = 2Δ/kBTc = 4.2 ± 0.2 inherit this uncertainty.","section":"Supplement, Fig. S3; main text 'Based on our observations'"},{"comment":"The ZBC simulation in Fig. 3 is not an independent validation of the superconducting scenario. It assumes an s-wave Dynes density of states with Δ = 9 meV, Γ = (8 ± 2) meV, and RBCS = 4.2, and the latter directly fixes Tc through RBCS = 2Δ/kBTc. The agreement between the simulated and measured ZBC curves is therefore built in by construction and cannot serve as a check of the claimed gap size or critical temperature. The simulation merely demonstrates that a strongly lifetime-broadened BCS DOS can reproduce a qualitatively similar ZBC trace; it does not distinguish superconductivity from other gap mechanisms.","section":"Fig. 3 and Supplement Eq. (S1)"},{"comment":"The identification of the observed dip as a superconducting gap is not established. The spectra show no coherence peaks, and the Discussion explicitly lists incoherent pair formation and proximity to a quantum critical point as alternative mechanisms that produce a gap with suppressed coherence peaks. The statement that 'the nonzero ZBC value can be attributed to the surface nature of the superconductivity' is an assumption, and the paper provides no vortex, field-dependence, Meissner, or transport data to support it. Without such evidence, the interpretation of the gap closing as 'the transition between the superconducting ground state and the normal state' is not forced by the data.","section":"Discussion paragraph on missing coherence peaks"},{"comment":"The uncertainty quoted for RBCS = 4.2 ± 0.2 does not propagate the systematic uncertainty in Δ. The paper itself states that FWHM/2 is 'almost certainly an underestimation of the real gap size' for reasonable Γ (Supplement Fig. S1), which means the true BCS ratio is a lower bound, not a measured value. In addition, the error bar neglects the excluded 50 K spectrum and the unknown effect of the linear-background subtraction. The comparison with weak-coupling values (Hg, MgB2) is therefore presented with a precision the data do not support.","section":"Main text, paragraph on BCS ratio"}],"minor_comments":[{"comment":"The abstract states 'a closing of the gap around Tc ≈ 45 K' while the body concludes Tc ≈ (50 ± 5) K; these two statements should be made consistent.","section":"Abstract and main text"},{"comment":"The methods section states set-point parameters for spectra are U_Bias = 50 mV and I_T = 500 pA, but the topography in Fig. 1(a) uses I_T = 50 pA; the difference should be clarified in the figure caption or text.","section":"Methods and Fig. 1 caption"},{"comment":"The annotation '2Δ = 18 mV' should use meV as the energy unit.","section":"Fig. 1(b)"},{"comment":"Reference [10] appears as a footnote-like note rather than a full citation; it should be converted to a proper reference with authors and journal.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper presents an interesting raw dataset, but the quantitative conclusions (Tc, RBCS) go beyond what the data can establish. I recommend asking the authors to either add decisive measurements (e.g., magnetic-field dependence, vortex imaging, or surface transport) or substantially soften the claims to a temperature-dependent spectral-gap observation. No issues of plagiarism or dual submission are apparent."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The temperature-dependent STS data on t-PtBi2—a ~9 meV dip in dI/dU that fills in between 8 K and 45 K—is genuinely new and directly visible in the raw spectra. But the paper's headline claim, that this is surface superconductivity with Tc ≈ 50 K and a BCS ratio of 4.2, rests on an interpretation that the data does not force.\n\nWhat the paper does well: the temperature series is not in the earlier isothermal work (refs. 1 and 4), and the comparison to thermally broadened 8 K spectra makes a real point—the dip closes faster than ordinary thermal broadening. The supplement is unusually transparent: it explains the FWHM-based gap estimate, acknowledges the absence of coherence peaks, and shows the excluded 50 K spectrum. The authors do not hide their uncertainties.\n\nThe soft spots are real but not fatal. The central issue is that the dip is never shown to be a superconducting gap. The paper itself mentions fluctuations and quantum criticality as sources of a gap with suppressed coherence peaks, but offers no measurement—vortices, surface transport, Meissner response—to rule those out. The Fig. 3 ZBC simulation is not an independent check: it assumes a Dynes DOS with Δ = 9 meV, Γ = 8 ± 2 meV, and RBCS = 4.2, so the claimed BCS ratio is built in. And since the 50 K spectrum was excluded because of a junction change, the Tc ≈ 50 ± 5 K estimate is an extrapolation from spectra that still have a finite dip. The uncertainty on Tc also understates the effect of the arbitrary Γ.\n\nThat said, the paper is honest and the raw data is worth reporting. If the authors re-frame the conclusion as a temperature-dependent gap of unknown origin, with the high-Tc interpretation as a hypothesis, the experimental contribution stands. As written, the abstract and conclusion overstate what is established.\n\nWho this is for: people working on t-PtBi2 and topological surface superconductivity, and STS groups interested in pseudogap versus superconducting-gap distinctions. I'd send it to review—a serious referee can push for the missing discriminating measurements and for the raw spectra. But I wouldn't cite it in my own work until the superconducting identification is strengthened.","headline":"The temperature-dependent STS data are a real experimental asset, but the paper overreaches by asserting Tc ≈ 50 K surface superconductivity without excluding pseudogap alternatives.","tokens_in":9026,"tokens_out":5569,"would_cite":false,"duration_ms":43980,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper reports that the Type-A surface of the Weyl semimetal t-PtBi2 carries a superconducting gap near 9 meV that fills in and closes only around 45–50 K, far above the roughly 1 K bulk transition, and it uses this temperature…","keywords":["surface superconductivity","t-PtBi2","Weyl semimetal","scanning tunneling spectroscopy","topological superconductor","Fermi arcs","BCS ratio","Dynes lifetime broadening"],"falsifier":"Measure the magnetic response of a cleaved t-PtBi2 surface between 10 K and 45 K: observing a Meissner expulsion or a vortex lattice would confirm superconductivity, while finding no diamagnetic response while the spectral dip persists would falsify the superconducting interpretation of the gap.","tokens_in":7816,"feed_emoji":"🔬","tokens_out":8828,"duration_ms":94420,"temperature":0.7,"pith_summary":"The paper reports scanning tunneling spectroscopy of cleaved t-PtBi2 from 8 K to 45 K on the Type-A, decorated-honeycomb surface. It finds a zero-bias depression in the differential conductance whose half width at half minimum is (9.0 ± 0.5) meV at 8 K, and which progressively fills in on warming, with almost no dip left at 45 K and a gapless, step-like spectrum at 50 K. The authors interpret this as a superconducting gap intrinsic to the surface, closing at a surface critical temperature near (50 ± 5) K, with a BCS ratio of 4.2 ± 0.2. If correct, this is high-temperature surface superconductivity in a topological semimetal while the bulk remains normal down to about 1 K, making t-PtBi2 a candidate intrinsic topological superconductor.","feed_headline":"Surface of t-PtBi2 superconducts to near 50 K","feed_subtitle":"Scanning tunneling spectra show a 9-meV gap closing near 45 K, about 50 times the bulk transition temperature.","key_machinery":"The carrying object is the zero-bias dip in the normalized dI/dU spectrum measured with scanning tunneling spectroscopy. With coherence peaks absent, the gap size is estimated from the full width at half minimum rather than from a BCS fit, and the temperature evolution is tracked through the zero-bias conductance (ZBC). The authors compare the measured ZBC(T) to simulations of an s-wave Dynes density of states with Δ = 9 meV and Γ = (8 ± 2) meV, and they overlay the thermally broadened 8 K spectrum at each temperature to show that the observed closing is not simple Fermi-edge broadening. The Dynes lifetime parameter plays a central role: it suppresses coherence peaks, makes the half width at half minimum a lower bound on the true gap, and lets the simulated ZBC(T) curves reproduce the data on a comparable numerical scale.","core_discovery":"The central claim is that surface superconductivity in t-PtBi2 is a separate, much higher-temperature order than the bulk transition: a large gap of about 9 meV persists above 40 K and closes only near 45–50 K. Because the spectra lack coherence peaks, the authors do not fit a BCS density of states; instead they take the half width at half minimum as a lower bound on the gap and test the temperature dependence of the zero-bias conductance against Dynes-broadened s-wave simulations with Δ = 9 meV, Γ = (8 ± 2) meV, and a BCS ratio of 4.2. The measured zero-bias conductance follows the simulated curve qualitatively, and thermal broadening alone cannot reproduce the observed closing. The paper therefore concludes that the surface critical temperature is about (50 ± 5) K and that the pairing is weak-to-moderate coupling, with 2Δ/kBTc = 4.2 ± 0.2.","pith_inferences":["Going beyond the paper, a field-dependent tunneling study is the natural next test: if the 9 meV dip is a superconducting gap, an in-plane field should suppress it or generate vortices, whereas a correlation gap would show a different field response.","Because STS averages over momenta while the Fermi-arc order is localized away from the Γ point, tunneling-matrix-element effects could make the apparent gap and its temperature dependence vary between tips; the paper notes this possibility but does not quantify it.","If the surface superconductivity is confined to the top layer, its Tc could be tunable by electrostatic gating or stacking variations; the paper mentions known two-dimensional Tc shifts in thin films but does not test them here."],"forward_implications":["The surface of t-PtBi2 would exhibit a superconducting state that sets in near 50 K while the bulk stays normal until about 1 K, implying the surface order is effectively two-dimensional and decoupled from bulk superconductivity.","The derived BCS ratio of 4.2 ± 0.2 places the surface pairing in the weak-to-moderate coupling range, closer to conventional superconductors than to strongly coupled cuprates.","Because lifetime broadening Γ is comparable to the gap, the true gap could be larger than the estimated 9 meV, so a precise gap determination is needed to firm up Tc and the coupling classification.","The coexistence of a high surface Tc with topologically nontrivial Fermi arcs makes t-PtBi2 a practical platform in which to search for zero-bias Majorana modes at the surface."],"supporting_citations":[{"why":"ARPES observation of a superconducting gap at the Fermi arcs of t-PtBi2, linking the surface gap to topological surface states.","marker":"[1]"},{"why":"Earlier STS work on t-PtBi2 surfaces reporting large gaps up to about 20 meV at 5 K and introducing the surface-termination classification used here.","marker":"[4]"},{"why":"Crystal growth and bulk characterization establishing that bulk superconductivity appears only near 0.6–1.1 K, the baseline against which the high surface Tc is contrasted.","marker":"[5]"},{"why":"Dynes density-of-states formula used to simulate lifetime-broadened spectra and the zero-bias conductance temperature curves.","marker":"[9]"},{"why":"Supplemental gap simulations and the 50 K spectrum used to argue that the gap is fully closed above 45 K.","marker":"[11]"},{"why":"Standard superconductivity reference supplying the weak-coupling BCS ratio of 3.54 and comparison values for computing the derived BCS ratio of 4.2.","marker":"[12]"}],"fun_headline_variants":["t-PtBi2 surface superconductivity persists to 45 K","Surface gap of 9 meV closes near 45 K in t-PtBi2","t-PtBi2 shows surface superconductivity up to 45 K","Distinct surface order in t-PtBi2 survives to 45 K"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the zero-bias dip in the tunneling spectrum is a genuine superconducting gap; if it is instead a pseudogap, density-wave gap, or fluctuation gap, then the Tc ≈ 50 K estimate read off its disappearance has no support.","fun_headline_variants_meta":{"raw":{"variants":["t-PtBi2 surface superconductivity persists to 45 K","Surface gap of 9 meV closes near 45 K in t-PtBi2","t-PtBi2 shows surface superconductivity up to 45 K","Distinct surface order in t-PtBi2 survives to 45 K"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000839,"raw_usage":{"total_tokens":3612,"prompt_tokens":856,"completion_tokens":2756,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":472,"completion_tokens_details":{"reasoning_tokens":2673}},"tokens_in":472,"tokens_out":2756,"duration_ms":27191,"temperature":1.0,"reasoning_tokens":2673,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T17:40:13.845885+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the magnetic response of a cleaved t-PtBi2 surface between 10 K and 45 K: observing a Meissner expulsion or a vortex lattice would confirm superconductivity, while finding no diamagnetic response while the spectral dip persists would falsify the superconducting interpretation of the gap.","supporting_citations":[{"cited_title":"Kuibarov, O","cited_arxiv_id":null,"evidence_quote":"ARPES observation of a superconducting gap at the Fermi arcs of t-PtBi2, linking the surface gap to topological surface states."},{"cited_title":"Schimmel, Y","cited_arxiv_id":null,"evidence_quote":"Earlier STS work on t-PtBi2 surfaces reporting large gaps up to about 20 meV at 5 K and introducing the surface-termination classification used here."},{"cited_title":"Hoffmann, S","cited_arxiv_id":null,"evidence_quote":"Crystal growth and bulk characterization establishing that bulk superconductivity appears only near 0.6–1.1 K, the baseline against which the high surface Tc is contrasted."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Dynes density-of-states formula used to simulate lifetime-broadened spectra and the zero-bias conductance temperature curves."},{"cited_title":"Such a high transition temperature is to date only known for unconventional superconductors like cuprates and Fe-based superconductors [12, 13]","cited_arxiv_id":null,"evidence_quote":"Supplemental gap simulations and the 50 K spectrum used to argue that the gap is fully closed above 45 K."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Standard superconductivity reference supplying the weak-coupling BCS ratio of 3.54 and comparison values for computing the derived BCS ratio of 4.2."}],"review_version":1}